| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Centre de recherche en biologie forestière, Univ. Laval, Sainte-Foy, QC, G1K 7P4 Canada
b Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Center, 1055 du P.E.P.S., P.O. Box 3800, Sainte-Foy, Qc, G1V 4C7 Canada
c Dep. of Soil Science, Univ. of Saskatchewan, 51 Campus Dr., Saskatoon, SK S7N 5A8 Canada
* Corresponding author (dpare{at}cfl.forestry.ca)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BS, base saturation • CEC, cation exchange capacity • SOH, stem-only harvesting • WTH, whole-tree harvesting
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Many simulation models of forest growth and timber yield, such as FORMAN, used in central and eastern Canada, and SYLVA II, applied in Quebec, assume that soil nutrient resources are in equilibrium with harvesting practices, that is, soil fertility is stable. However, clear-felling influences nutrient cycling in forest ecosystems. Both theoretical (e.g., Paré et al., 2002; Wei et al., 2000; Bhatti et al., 1998; Freedman et al., 1986; Sachs and Sollins, 1986; Aber et al., 1978; Weetman and Webber, 1972) and empirical (e.g., Bélanger et al., 2003; Egnell and Valinger, 2003; Johnson and Todd, 1998; Titus et al., 1998; Olsson et al., 1996a; Hendrickson et al., 1987; Nykvist and Rosén, 1985; Adams and Boyle, 1982; Johnson et al., 1982) studies have shown that the intensity of harvesting can have significant mid-term effects on soil nutrient resources and possibly effects on long-term site productivity.
For example, Olsson et al. (1996a) observed, 15 yr after harvest of boreal coniferous stands in Sweden, a reduction between 8 and 19% in base saturation (BS) in the forest floor following WTH compared with SOH, as well as a higher C/N ratio, which may negatively influence potential N mineralization in the soil. After 24 yr of growth, the stands regenerated after WTH had 20% less wood biomass than stands regenerated in the SOH plots (Egnell and Valinger, 2003). These results therefore challenge the assumption made in tree growth models that soil nutrient resources are in steady state with harvesting operations, particularly WTH.
The biogeochemical model developed by Paré et al. (2002) suggested that the impacts of WTH on ecosystem nutrient balance are site-specific, depending on the type of parent material and species harvested. To limit nutrient depletion, the model suggests that WTH of nutrient demanding tree species (e.g., balsam fir) should be avoided on shallow soils and glaciofluvial sands.
To validate and fine-tune these recommendations, we wanted to generate empirical information about the impacts of different harvesting intensities for an array of sites with contrasting soil characteristics. Our objective was to compare, 15 to 20 yr after harvest, the effects of WTH and SOH on soil and foliar nutrient status for boreal coniferous stands. We chose this time-frame to allow the potential effect of logging residues on nutrient cycling, if any, to take place. To cover a wide array of conditions, our study was conducted in three types of stands characterized by various levels of nutrients in foliage and branches (balsam fir > black spruce > jack pine; Paré et al., 2002) as well as on soils of contrasting fertility.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sites in Haute-Mauricie are classified as a black spruce-feathermoss-ericaceous ecological type on a coarse-textured mesic parent material (Ministère des Ressources naturelles du Québec, 2000), whereas sites in Forêt Montmorency and Gaspésie are classified as a balsam fir–white birch ecological type on a medium textured parent material (Ministère des Ressources naturelles du Québec, 1999). Mean annual precipitation and air temperature in Haute-Mauricie are 1000 mm and 1°C, respectively (Ministère des Ressources naturelles du Québec, 2000). In Forêt Montmorency and Gaspésie, mean annual temperature is 0°C and mean annual precipitation is respectively 1400 and 1300 mm (Ministère des Ressources naturelles du Québec, 1999). Haute-Mauricie and Forêt Montmorency plots lie on granitic or granitic gneiss of the Canadian Shield, whereas Gaspésie plots are on soils developed from Appalachian schists. Soils in Gaspésie are typic Haplohumods developed from sandy loam till deposits (Soil Survey Staff, 2003). Both Haute-Mauricie and Forêt Montmorency soils are typic Haplorthods developed from loamy sand glaciofluvial deposits in the former region and loamy sand till in the latter region (Soil Survey Staff, 2003). All soils have a mor humus with a mean depth of 5, 8, and 2 cm in Haute-Mauricie, Forêt Montmorency, and Gaspésie, respectively. All soils also have an E horizon overlying the spodic B horizon, ranging from 2 to 10 cm in depth.
Experimental Design
The field experiment in the Haute-Mauricie region is a randomized complete block design, with two treatments and fourteen blocks (one replicate per block, n = 14). Forest composition, either black spruce or jack pine, was similar before and after felling within a block, but varied across blocks. Eleven blocks were scarified and replanted after felling, while the remaining three were regenerated naturally with no site preparation. Forêt Montmorency and Gaspésie each have a completely randomized design, with two treatments and four replicates (n = 4) in Gaspésie, and five replicates (n = 5) in Forêt Montmorency, with only one tree species, balsam fir, before and after harvesting. The treatments are (1) WTH, that is, felling and hauling of all above-stump biomass, and (2) SOH, where delimbing and topping occur on site and only the stem is hauled, which leaves all logging residues at the soil surface. In each region, replicates are located in different cutblocks, frequently defined by changes in forest operators and equipment, and are separated by at least 180 m. All experimental plots are 10 by 10 m.
Field Sampling
Soil sampling was performed in July and August 2001 in Haute-Mauricie and Gaspésie and in July 2003 in Forêt Montmorency. Within each plot, 10 locations were randomly sampled. Both the forest floor and the first 20 cm of the spodic B horizon (hereafter referred to as the mineral soil) were sampled volumetrically at each of these locations. Moss cover and large roots were removed from the forest floor samples. Moist samples were air-dried and sieved at 4 mm (forest floor) or at 2 mm (mineral soil).
In October 2002, foliage was sampled within each plot of the study sites on 10 dominant or codominant trees. Five samples of current-year needles and five samples of first-year needles were collected from the upper third of the canopy of selected trees. Foliage was oven-dried at 65°C for 72 h.
Chemical Analyses and Calculations
Soil pH in distilled water (pHwater) was determined with a glass electrode-calomel electrode system (pHM82, Standard pHMeter, Radiometer Copenhagen, Brønshøj, Denmark) in distilled water using a 1:4 and 1:2 soil to solution ratio for forest floor and mineral soil, respectively.
Soil exchangeable cations were extracted using an unbuffered BaCl2 solution (Hendershot et al., 1993) and determined by atomic absorption (Ca, Mg, Fe, Mn, and Al) and by atomic emission (Na and K) (5100PC Atomic absorption spectrophotometer, PerkinElmer, Wellesley, MA). Exchangeable H+ in the forest floor was assessed from pHwater using the model developed by Bélanger et al. (2006). They showed that while the contribution of exchangeable H+ to the CEC was high in the humus layer, it was relatively small and difficult to predict in the upper spodic B horizons of Quebec and thus it was not considered for these horizons. Cation exchange capacity was calculated as the sum of exchangeable cations (including H+ for the forest floor). Base saturation was calculated as the contribution of Ca, Mg, K, and Na to CEC.
Ground soil samples were Kjeldahl-digested and analyzed with the Quickchem Method for total N (Zellweger Analytic, Inc. Lachat Instruments Division, Milwaukee, WI). Organic C content was determined by loss-on-ignition on ground forest floor samples (Nelson and Sommers, 1996), and by potassium dichromate digestion and subsequent ferrous sulfate titration (Yeomans and Bremner, 1988) for ground mineral soil samples.
In each plot, two samples of the mineral soil were selected randomly for particle-size distribution and total chemical composition. Particle-size distribution was determined by sieving and by sedimentation analysis using a laser particle sizer (Analysette 22 compact, Fritsch GmbH & Co KG, Welden, Germany) on NaOCl-pretreated samples. Elemental composition was determined on 32-mm diam. fused beads prepared from a 1:5 soil/lithium tetraborate mixture using an automated x-ray fluorescence spectrometer system (Philips PW2440 4kW, Panalytical, Almelo, The Netherlands) with a Rhodium 60-kV end window x-ray tube.
Foliage was digested with a H2SO4–H2O2 solution (Parkinson and Allen, 1975) and analyzed for Ca, Mg, and K concentrations by atomic emission. Needle N concentrations were determined with the Quickchem Method. Two subsamples of 100 needles were weighed for each needle sample to determine mass per needle. Nutrient content was obtained by multiplying concentrations by mass per needle.
Data Analyses
Means of the sampling points in each plot were used for all analyses. Analysis of variance (ANOVA) was used to test for treatment effects on soil and foliar variables. Due to differences in experimental design, the ANOVA for Haute-Mauricie was performed independently of those for Forêt Montmorency and Gaspésie.
For Haute-Mauricie, means for soil variables were tested using a two-way ANOVA model, with harvesting method and block as sources of variation. Species was not considered at that time: statistical analyses on soil data and including species as a source of variation showed no significant difference related to stand composition. For needles, the model was modified to take into account regenerating species (black spruce or jack pine) as a source of variation. Therefore, means were tested according to a nested model, with species, block (nested within species), harvesting method and species x harvesting method interaction as sources of variation. To take into account the low number of denominator degrees of freedom and the high heterogeneity of the studied variables, statistical significance was accepted at 
= 0.10. As argued by Peterman (1990), in environmental studies, the practical consequences of a Type II error, that is, failing to detect a difference which did occur in nature, and which is related to the level selected, may be more serious than the consequences of a Type I error, that is, detecting a difference which did not occur in nature.
Soil and foliar variables in Forêt Montmorency and Gaspésie were tested using a nested ANOVA model with location, harvesting method (nested within location) and location x harvesting method interaction as sources of variation. Location and harvesting method effects were considered statistically significant at = 0.10. However, location x harvesting method interaction was considered significant at = 0.30 as proposed by Milliken and Johnson (1984) in the case of multilocation trials.
ANOVAs were conducted using the PROC MIXED procedure in SAS v. 8.1 and the Satterthwaite approximation option was used to calculate the denominator degrees of freedom (Littell et al., 1996). Data were log-transformed to assure homogeneity of variance, except for pH and BS.
Vector diagnosis was used to visualize foliar concentration and content in an integrated graphic format. This technique simultaneously compares nutrient concentration, nutrient content, and plant or plant component biomass (in this case, needle biomass) in a vector nomogram (Salifu and Timmer, 2001). Whole-tree harvesting was used as the reference treatment. Means of concentrations and contents were normalized within each species x location category, by dividing the value by the corresponding value for the WTH treatment and then by multiplying it by 100. Nutritional interpretations of directional shifts were made according to the framework and terminology (dilution, sufficiency, deficiency, luxury consumption, excess, and depletion) in Salifu and Timmer (2001) (Fig. 1 ).
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
|
Harvesting methods did not have a significant effect on BS (Tables 2 and 3). Also, no statistically significant difference in pHwater was found between harvesting methods.
Molar ratios of exchangeable base cations (Ca, K, and Mg) to exchangeable Al measured in forest floor were highest in Gaspésie and lowest in Haute-Mauricie (Table 2). Opposite results were found for the mineral soil. These ratios were not affected by harvesting methods at either soil depth in Gaspésie and Haute-Mauricie (Tables 3 and 4). However, in the mineral soil, there was a significant interaction between Location (i.e., Forêt Montmorency and Gaspésie) and Harvesting method (p = 0.068). This was also the case for BS (p = 0.056), exchangeable K (p = 0.277), and exchangeable Mg (p = 0.074) in the mineral soil. In all cases of significant Location x Harvesting method interaction, the direction of the observed response in Forêt Montmorency was opposite to that observed in Gaspésie and Haute-Mauricie.
Soil Total Nitrogen and Organic Carbon
For both soil layers, the Gaspésie region presented the highest values of total N and organic C, while values were lowest in Haute-Mauricie (Table 2). In Haute-Mauricie, amounts of organic matter in the mineral soil were very low, that is, approximately 1% of organic C and 0.05% of total N. Harvesting methods did not significantly influence total N concentration at either soil depth (Tables 3 and 4). In all regions, organic C concentration tended to be higher in the forest floor under SOH, albeit no statistical difference was detected. This variable was highly heterogeneous (Table 2). For the mineral soil, organic C in Haute-Mauricie was significantly affected by harvesting method (p = 0.038), with SOH having the highest concentration. No differences were detected between harvesting methods at the other two sites.
Needle Mass and Chemical Composition
The ANOVA results showed that in Haute-Mauricie, jack pine had significantly heavier needles in the SOH plots (current-year needles: p = 0.060, first-year needles: p = 0.001) (Tables 5 and 6). The same trend was observed for black spruce but was significant in current-year needles only (current-year needles: p = 0.060, first-year needles: p = 0.346). For both jack pine and black spruce in Haute-Mauricie, Mg concentrations were significantly higher after SOH in current-year (p = 0.065) and first-year (p = 0.017) needles (Tables 5 and 6). Phosphorus concentrations were similar between harvesting methods in black spruce, but were 8% higher in current-year needles of jack pine grown after SOH (p = 0.022).
|
|
Impacts of harvesting method on foliar nutrients of balsam fir were not as clear because of contrasting effects of the treatments in the Forêt Montmorency and Gaspésie sites. For most nutrients, this caused a significant interaction between location x harvesting method (p < 0.3) (Table 7).
|
Balsam fir in WTH plots had significantly higher nutrient content in needles of both ages in Forêt Montmorency, except for Ca content in first-year needles, which was similar between harvesting methods (Tables 5 and 7). In contrast, balsam fir in Gaspésie had significantly higher (p = 0.028) Ca content in current-year needles after SOH, but no harvesting method effect was observed for other nutrients.
These results are more easily visualized with vector analysis (Fig. 2 ). The amplitude of the effect, expressed by vector length, shows that differences between harvesting methods for all nutrients were more pronounced in first-year compared with current-year needles for black spruce and jack pine in Haute-Mauricie. Contrasting results were obtained for balsam fir in Gaspésie: vector length was marginal for all nutrients in first-year needles, except for Ca.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the present study, harvesting intensity had an influence on CEC. This may be due to differential effects on pH, as well as organic matter quantity and/or quality. In most soils, CEC is pH-dependent and thus increases with increasing pH (Stevenson, 1994). In contrast, we observed that CEC was higher under SOH while pH was roughly similar. Other studies (Staaf and Olsson, 1991; Nykvist and Rosén, 1985) showed significant reduction in soil pH under WTH compared with SOH.
The high production and persistence of organic acids in soils of cold-temperate climates such as our study sites may explain part of this difference (Bruckert, 1986). In our study, soil pH in SOH plots was possibly driven by organic acids produced in logging residues and leached into soil.
Since most of the CEC of Spodosols is attributed to organic matter (Stevenson, 1994), higher CEC after SOH may result from increased soil organic matter. This may be the case in the mineral soil in Haute-Mauricie, where organic C concentrations were significantly higher under SOH. Leaving branches and foliage on site may have caused a translocation of organic matter to lower horizons, which can be favored by the coarse texture of the parent material (Olsson et al., 1996b). We detected no such effect on organic C in Gaspésie and Forêt Montmorency. The meta-analysis of Johnson and Curtis (2001) suggests variable effects of harvesting intensity on mineral soil C in coniferous forests; whereas some studies demonstrate a net positive effect of leaving debris on site on mineral soil C pools, others show little or no difference between WTH and SOH. Ussiri and Johnson (2004) observed that mineral soils with high organic matter content inhibit sorption of new C. Therefore, the lack of response in mineral soil C to residue management in Gaspésie and Forêt Montmorency may be due to preexisting high concentrations of organic C, compared with low levels measured in Haute-Mauricie.
It is also possible that soil organic matter quality differed under WTH and SOH (e.g., fulvic/humic acid ratio, nature and availability of functional groups), thereby influencing soil exchange properties (Johnson et al., 1991, 1997). Inputs from slash decomposition may modify the chemical nature of organic C in soil (Mathers et al., 2003). Whether these modifications are linked to greater CEC of SOH soils remains to be tested.
Differences between Nutrients
The presence of logging residues on site resulted in increased forest floor exchangeable Ca concentrations, an effect noted in other studies (e.g., Johnson and Todd, 1998; Olsson et al., 1996a). This response occurred under a wide array of soil conditions. Due to its low mobility in soil and slow release by decomposition (Edmonds, 1987), Ca contained in woody debris can be easily captured by the forest floor, thereby creating a marked difference with slash-free harvested sites.
Enhanced Ca availability in the forest floor after SOH was reflected in higher foliar Ca concentrations in the regenerated stands. This was true across all regions, but this effect was particularly evident in Gaspésie. Gaspésie soils have very low elemental Ca concentrations (as also observed by Bélanger et al., 2004) and thus trees could be limited in regard to Ca resources. Stem-only harvesting appears to have alleviated the Ca nutritional deficiency observed in stands regenerated after WTH. Detrimental effects of WTH on Ca supplies have been documented in a large number of studies (e.g., Freedman et al. [1986] in central Nova Scotia; Johnson et al. (1982) in eastern Tennessee; Weetman and Webber, 1972 in northern and southern Quebec).
Haute-Mauricie soils have low elemental Mg typical of many sites on the Canadian Shield (see Bernier and Brazeau, 1988; Courchesne et al., 1996; Houle et al., 1997) and consequently, the benefits of SOH on foliar nutrition were large for Mg. However, contrary to Ca, effects on foliar Mg could not be linked to observable differences in soil exchangeable Mg pools. This may be due to the fact that with the same number of samples, statistically significant effects are more likely to be detected in foliage, since foliage has less variable nutrient concentrations than surface soil. Moreover, the chemistry of upper forest soil layers, particularly the forest floor, is controlled by litterfall chemistry. Since litterfall and all types of biological material appear to have relatively constant nutrient ratios (e.g., Knecht and Göransson, 2004), the signal of the low Mg and Ca availability in upper soil layers is likely weaker than that of the elemental chemistry of the parent material. In this respect, parent material composition may be an important predictor of tree Ca and Mg nutrition, especially under stressful conditions where the biogeochemical cycle of elements is disrupted. Thus a model that considers soil mineral weathering of the parent material would likely be useful in predicting stand nutrition and productivity. Interestingly, Bailey et al. (2004) also considered the availability of Ca and Mg as weathering products in soils as a good indicator of sugar maple (Acer saccharum Marsh.) susceptibility to decline.
Olsson et al. (1996b) found that the effects of logging residue management on exchangeable K are quite different from effects on divalent base cations. This is reflected in foliar nutrition of regenerating stands (Olsson et al., 2000); K released by mineralization of logging debris is apparently rapidly lost through leaching due to its high mobility in soil and is therefore not retained by the growing stand. Similar pattern of K cycling was also reported by Goulding and Stevens (1988), Fahey et al. (1991), and Proe et al. (1999). In our study, logging residues left on site in Haute-Mauricie and Gaspésie after SOH seemed to have created a K flux through the soil profile capable of modifying the equilibrium of the mineral soil exchange complex (Bélanger et al., 2003). Differences between harvesting methods were not statistically significant in Gaspésie, but this is possibly attributable to a low number of replicates. In accordance with the conclusions of Proe et al. (1999), it seems that mineralization of logging residues added little K easily available for tree uptake because it was leached to deeper horizons or lost in ground and surface waters.
Fifteen to twenty years after disturbance, impacts of harvesting methods on foliar N were less apparent than effects on divalent base cations. This result was similarly observed by Olsson et al. (2000) in Picea abies (L.) Karst. and Pinus sylvestris L. stands. These latter authors observed that logging residues significantly increased foliar N concentrations of the regenerated stands in earlier stages of growth (8–10 yr), but this effect decreased with time. Enhanced N nutrition with SOH could not be explained by a greater total N pool (Olsson et al., 1996b). However, at the local scale, total N may not provide an accurate index of N availability (Binkley and Hart, 1989). We evaluated potentially mineralizable N for WTH and SOH in the Haute-Mauricie region using anaerobic incubation of soil for 14 d at 30°C, but results were inconclusive. Analysis of the dynamics of decomposition made by Hyvönen et al. (2000) showed that N release from logging debris in boreal ecosystems may continue well past 15 yr after disturbance. However, N fluxes may be too low by then to be detected by short-term anaerobic incubation analyses and to significantly influence tree nutrition.
Differences between Species
The three species studied—jack pine, black spruce, and balsam fir—reacted differently to harvesting intensity. Jack pine appeared to be more opportunistic than black spruce and apparently, was able to use the flow of nutrients made available by decomposition of logging residues in SOH systems. This result is not an artifact caused by differences in soil fertility between jack pine and black spruce stands, since no significant differences in mineral soil C and nutrient concentrations were detected in relation to stand composition (data not shown). Jack pine is a pioneer species with a relatively fast growth rate, especially during juvenile growth. According to Vasiliauskas and Chen (2002), the time required to reach a height of 1.3 m is 18 yr for black spruce and 8 yr for jack pine regenerating after fire in Ontario. Jack pine develops a root system that penetrates deep in the soil (Burns and Honkala, 1990) and does not keep its needles more than 2 to 3 yr. On the other hand, black spruce is often a late-successional species, has a shallow rooting habit, grows slowly (Burns and Honkala, 1990) and has a leaf longevity that can be 7 yr or more. According to the gradient of ecological functional traits described by Diaz et al. (2004), jack pine attributes are closer to those of the "acquisitive type" which is characterized by rapid acquisition of resources, whereas black spruce is more a "conservative type." Our results indicated that the species recolonizing the site might, to a certain extent, determine the effect induced by WTH.
There also seems to be a difference in nutrient retranslocation patterns between balsam fir, jack pine, and black spruce. Impacts of harvesting intensity were stronger in first-year needles of black spruce and jack pine, but these were clearer in balsam fir current-year needles. This may indicate that the two former species transfer nutrient reserves to current-year foliar tissues, making nutritional deficiencies more obvious in older needles. Balsam fir may be less efficient at nutrient retranslocation than jack pine and black spruce.
The Case of Forêt Montmorency
Contrasting results obtained in the Forêt Montmorency region compared with Haute-Mauricie and Gaspésie prompted us to investigate the design in this region in more detail. Contrary to the latter two regions, where plots are on relatively flat terrain, plots in Forêt Montmorency present slopes ranging from 5 to 24%. After verification, it appeared that WTH plots were generally facing south, while SOH plots faced north. Estimation of solar radiation for all plots in Forêt Montmorency using the method described by Nikolov and Zeller (1992) showed that the average sum of radiation received during the growing season (May to September) is 2649 and 2561 MJ m–2 in WTH and SOH plots, respectively. An ANOVA performed to assess the difference between these values produced a p of 0.086. We hypothesize that the difference in solar radiation received by plots of WTH and SOH explains at least part of the contradictory responses of foliar nutrition and soil properties related to harvesting intensity at Forêt Montmorency. Solar radiation could not be used as a covariable in the analyses because relationships with soil and foliar parameters were nonlinear.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Improved needle divalent base cation status after SOH appears to be linked to elemental composition of the parent material: stands were grown either on low elemental Mg soils or low Ca soils. These sites presumably receive small amounts of these nutrients from mineral soil weathering and in turn, nutrition is unbalanced and limited in regard to these nutrients. Interestingly, the total elemental content of mineral soil horizons seemed to be a better indicator of sensitive conditions in boreal forests than exchangeable elemental concentrations. One possible explanation is that in surface soil layers, nutrient concentrations are largely controlled by litterfall, which has relatively well-balanced nutrient ratios. When trees need to rely on soil reserves, intrinsically poor soil or soil with imbalanced nutrient proportions may quickly cause dysfunctional tree nutrition.
Finally, tree species with attributes that allow rapid acquisition of resources, like jack pine, reacted more strongly to the impact of harvesting intensity on the soil nutrient status. We conclude that selecting tree species for reforestation of harvested sites can influence the outcome of WTH impacts. Conversely, differences in nutrient export associated with the various harvested species do not appear to be a major factor contributing to the observed effects of harvesting method.
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 19, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
